2 messages in com.mysql.lists.plusplusRe: MySQL C++ API
FromSent OnAttachments
Michel Stam02 Mar 2001 09:46 
Johann Gunnar Oskarsson02 Mar 2001 11:26 
Subject:Re: MySQL C++ API
From:Johann Gunnar Oskarsson (joha@firmanet.is)
Date:03/02/2001 11:26:05 AM
List:com.mysql.lists.plusplus

Hi,

My philosophy is: Patch first, ask questions later.

That way, I believe, I was the first to get MySQL++ to work on Redhat 7.0. It was a hack, it was ugly, but it worked ... as long as I used the correct constructor for the Connection class...

I then automate the untaring, patching, making, installing process. Which means I have to distribute the libraries too, along my software, but that hasn't been a problem so far.

Johann

On Friday 02 March 2001 17:46, you wrote:

Hello,

I've been writing some software based on the MySQL database and the MySQL C++ client. The project works, so no problems there.

However, while finalizing the project and trying to solve some security issues, I stumbled upon something which I believe I cannot solve myself.

What I'm trying to accomplish is get a connection to the mysql database which is SSL encrypted. The mysql C client API has some provisions for this.

To specify the paths to the certificates however, you need to use the mysql_ssl_set( ) and mysql_ssl_clear( ) functions before specifying through client flags that you wish to use the SSL connection feature.

I think I can work around the flags by using the real_connect member of the Connection( ) class, but I cannot manipulate the C clients' MYSQL structure directly, since it's a private member of this class. This is required for specifying the certificates. I'm not exactly sure yet as to whether these are paths to the files or the contents of these files themselves. I assume the former.

I thought of deriving a class from the Connection class to get around this problem, but as I mentioned before, the private member does not allow this.

Am I looking correctly at this ? If so, will there be a version of the C++ client which has this fixed ? Or is it possible to submit a patch to you which solves this and have it included in the mainstream C++ client distribution ?

Thanks in advance for any help.

Best regards,

Michel Stam